BACKGROUND: A behaviour change campaign is unlikely to be effective if its intervention is not carefully designed. While numerous frameworks are widely used to develop and evaluate interventions, the steps detailing how to create an intervention are not as clear because the process of linking behaviour analysis to the intervention design is seldom discussed. We document the application of the Behaviour Centred Design (BCD) approach to the development of an intervention to improve hand hygiene (HH) rates among nurses' hospital units in the USA. METHODS: Intervention development is divided into the first three steps of the BCD approach: Assess, Build, and Create. The Assess step centres on understanding the target behaviour. The Build step expands the knowledge of the target behaviour and population through formative research which leads to a creative brief that explains the focus of the intervention. In the Create step, the creative brief guides the intervention design. RESULTS: Drawing from the main findings of the Asses and Build steps, a focal insight was developed positing that nurses can rediscover the meaning and purpose of their role as a nurse and thus as a caregiver by practicing HH; in the process of cleaning their hands, nurses are living up to their ideal nurse-self. The focal insight was linked linguistically into a theory and change. The outcome was a simple intervention, called the Mainspring Intervention, which consisted of three major parts: a self-affirmation exercise to reduce defensiveness, a message that challenged nurses' perceptions about their HH practice, and an implementation intention activity to help nurses link HH behaviour to a cue. CONCLUSIONS: We detailed the creation of an original HH intervention that used the BCD approach. The intervention is relatively simple compared to most HH initiatives in the literature, both in terms of having relatively few components to the intervention and relatively easy field implementation. This intervention will allow us to test how specific psychological processes contribute to the problem of low HH rates, how our proposed intervention changes these processes in the hospital setting, and how the expected change in nurses' cognition transforms over time because of the intervention.